Waterman, D. #1986#. A Guide to Expert Systems. Addison Wesley.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....barrier , i.e. the fundamental difficulty that we all have in efficiently and effectively surveying and eliciting our own knowledge. Waterman describes this as knowledge engineering paradox : the more competent domain experts become, the less able they are to make explicit their knowledge [Waterman86]. We infer from this, that a tool based on the retrospective approach to knowledge survey would not be used. In fact, if knowledge and experience consist of what the knowledge worker constructs as proposes the radical constructivist approach it follows that she should primarily focus on the ....

Donald A. Waterman, A Guide to Expert Systems, Addison-Wesley, Reading, MA, 1986.

....the module ViewBuilder) in a form that is likely to be useful during execution. There is no way of validating these other than trial and error, which we claim is a fundamental consequence of real world reasoning. It is inevitable that part of the Knowledge Acquisition effort (see for example, [Waterman]) will be devoted to creating rules or heuristics that apply in appropriate contexts. Next we define two levels of relevance. The first relevance level applies to formulae which refer to the same key constants as the question QN. For example, if QN is the question, Is Richard Nixon a ....

D. A. Waterman, A guide to Expert Systems, Addison-Wesley, Reading, MA. (1986).

....of an expert system is the corpus of knowledge that accumulates during system building. In expert system design, the main role 42 Expert Knowledge Engineer Expert System Queries, problems Answers, solutions Strategies, rules of thumb, domain rules Figure 4 1: Knowledge Engineering, from [38] players are the expert system, the domain expert, the AI consultants, the knowledge engineer, the expert system building tools, and the user. Their relationships to each other are shown in Figure 4 2. The domain expert is a knowledgeable person with a reputation for producing right results in a ....

.... 43 ToolBuilder Expert System Building Tool Domain Expert Knowledge Engineer Expert System Clerical Staff End user Extends and tests Adds Data Builds, refines, and tests Builds Uses Interviews AI Consultants Figure 4 2: Roles in Expert System Building, adapted from [38] ffl A rule based representation uses the common statement IF . THEN . The matching of rule IF portions to the facts can produce inference actions. Rules provide a natural way to describe problem solving strategies. We can think about a scenario when we see a doctor and tell him what the ....

[Article contains additional citation context not shown here]

D. A. Waterman. A Guide to Expert Systems. Addison-Wesley, 1985.

....available. Hence, it is a major premise of this dissertation that learning should involve both examples and precepts that constitute prior knowledge. Much effort has recently been expended in understanding the sources and use of prior knowledge in learning [DES93] The strong knowledge principle [WAT86], early work on bias [MIT80] and the scarcity of interesting positive theoretical results [VAN94] suggest the difficulty of learning without sufficient a priori knowledge. That is, examples should indeed be augmented by prior (domain) knowledge. Such knowledge can be obtained from an expert or ....

....extensions. Many of these logics have been successfully implemented in artificial systems, such as the logic programming language PROLOG and a wide variety of expert systems (e.g. MYCIN for medical diagnosis, PROSPECTOR for geological prospection) Description of several systems and logics are in [BRE91, COR84, GIN87, LUK90, WAT86, WEB81]. Analogical, or similarity based reasoning can beneficially be used to complement deductive reasoning. In particular, when no rule exists that exactly matches a new situation, analogy may be used to attain a result. Analogical reasoning typically uses similarity with known situations to make ....

[Article contains additional citation context not shown here]

Waterman, D.A. (1986). A Guide to Expert Systems. Addison Wesley.

....1. 2 Normative expert systems in medicine Expert systems arose in the 70 s in the field of artificial intelligence as computer programs that, like human experts, possess a deep knowledge about a narrow domain, which allows them to solve problems by reasoning and explaining this reasoning [5, 41]. A great proportion of the most famous expert systems were built as diagnosis assistants and therapy advisors in different medical areas (MYCIN, PIP, CENTAUR, INTERNIST, ONCOCIN, etc. and almost all of them based their reasoning totally or partially on IF THEN rules, which, combined with frames ....

D. A. Waterman. A Guide to Expert Systems. Addison-Wesley, Reading, MA, 1986.

.... these attitudes, operators for adding mental attitudes, the cognitive frame, the language, control procedures, social interaction propensities, and principles and mechanisms for generating inter agent social behaviour (Carley 1989; Davis and Lenat 1980; Dennet 1987; Hayes Roth et al. 1983; Waterman 1986). The cognitive structure changes as the agent interacts with its physical and social environment and acquires new information. This, in turn, affects the agent s behaviour and its interaction with the environment (Carley 1989) The two main components of each negotiating agent s cognitive ....

Waterman, D. A. (1986). A Guide to Expert Systems, Reading, MA: Addison-Wesley Publishing Co.

....In addition to an actual question and answer session, one can utilize other techniques such as filling in questionnaires or even observing the way that experts work in real situations. The resulting system is based on rules for different kind of situations and is called a rule based system. Wat86, Nik97] Rule based systems can also be built without a human expert. If a sufficiently large body of data has been collected about the decision making situation, machine learning techniques can be used to extract rules automatically from the data. For this purpose, different techniques have ....

Waterman D A: A guide to expert systems, Addison-Wesley, 1986.

....hypotheses. Hypothesis testing continues recursively until some hypothesis is false or all hypotheses are verified with real time observations, in which case the system confirms the original fault hypothesis. An example of a diagnostic monitor which incorporates both inference paradigms is REACTOR [Waterman, 1986], an experimental monitor for nuclear power plants. FALCON [Shirley 1987] Rowan, 1989] is a similar prototype rule based system that identifies the causes of process disturbances in chemical process plants. 38 One of the problems of early expert systems was their inability to record history ....

Waterman D. A., A Guide to Expert Systems, Addison-Wesley, ISBN: 0-20108-313-2, 1986

....of knowledge acquisition when constructing knowledge based systems. Conventional knowledge acquisition is a painstaking process a series of intense, systematic interviews between knowledge engineers and domain experts to explicate the domain experts heuristics [Feigenbaum and McCorduck, 1983; Waterman, 1986] It is very common for domain experts to have difficulty fully expressing their knowledge or methods at an abstract level, but it is usually much easier for them to give examples of solving specific problems in their domains. Inductive learning systems take these examples and induce general ....

D.A. Waterman, A Guide to Expert Systems, Wokingham, UK: Addison-Wesley.

....the knowledge base. Figure 2.1 from Giarratano Riley [31, page 3] illustrates the basic concept of a knowledge based expert system. Facts Expertise Expert System Figure 2.1 The Basic Concept of an Expert System Function. ii) Is an expert system same as a knowledge based system Waterman [62] defines expert system and knowledge based system as follows: Knowledge Base Inference Engine User Page 7 Expert system. A computer program that uses expert knowledge to attain high levels of performance in a narrow problem. These programs typically represent knowledge symbolically, ....

Waterman, Donald A., "A Guide to Expert Systems", Addison-Wesley Publishing Company, Inc., 1986.

.... extracting the knowledge abound. This sort of thinking has a flavour of getting to archetypal knowledge. When the knowledge the expert gives is not good enough ( the archetypes have not been reached) the tendency is to blame the expert as being unable to report on his deep mental processes(Waterman, 1986). We believe that there must be archetypes somewhere and if necessary our knowledge bases will progress through common sense till we come to the archetypes. None of this is to deny the utility of knowledge , the success of expert systems and the future importance of large common sense knowledge ....

Waterman, D. (1986). "A guide to expert systems." Reading, MA, Addison Wesley.

....of the system called the inference engine. The domain knowledge is gathered from human experts. In addition, the expert system must carry out the task at a high level of competent performance, using the shortcuts or tricks that human experts use to eliminate wasteful or unnecessary calculations [Waterman,85] The implication from the statements above is that the knowledge must captured and represented in forms capable of being manipulated by the expert system. However, there are many tasks that human beings can solve easily using what is called common sense reasoning without being able to express ....

Waterman, D. A., A Guide to Expert Systems, AddisonWesley, Reading, Mass., (1985). 209

....field of chemistry. The human experts used experiential (heuristic) knowledge as well as factual knowledge. The problemsolving process was procedural in nature, and thus was suitable for incorporation as a network of rules. All of these traits match well with developing an expert system solution [21]. Second, cost benefit analysis favored using an expert system approach. By using graduate student labor, the application cost the company little. Use of the ES greatly simplified the organization and sequencing of the package decision process, thus saving time (with fewer, shorter meetings) and ....

Waterman, D. A Guide to Expert Systems. Reading, MA: Addison-Wesley, 1986.

....in which the system s analysis or plan was constructed, and why it was right. It is desirable for any system to be able to explain its operation. In rule based expert systems, this is called an explanation facility. Typically, it involves displaying the rules applied in arriving at a conclusion [Waterman, 1986] . Such an explanation, although it increases the users confidence in the system, can be opaque and difficult to follow. CBR systems have the advantage that by using cases, they can provide an explanation that is generally much easier to understand and more compelling than a chain of rules. If a ....

Waterman, Donald A. 1986. A guide to expert systems. Addison-Wesley Publishing Company.

....results of the 2 diagnostic system, which was implemented on real photolithography equipment in the Berkeley Microfabrication Laboratory. 2 Anatomy of a Diagnostic System A diagnostic system typically consists of three components: an inference engine, a knowledge base and a user interface [5] [6]. The function of the user interface is to provide an interface between the diagnostic system and the user, since all diagnostic systems are expected to work in conjunction with a human expert. The knowledge base is the component that contains the expert information needed to diagnose the problem. ....

....user, since all diagnostic systems are expected to work in conjunction with a human expert. The knowledge base is the component that contains the expert information needed to diagnose the problem. This information can be represented in several ways, such as a set of rules, frames, semantic nets [6], belief networks [7] or equipment models [8] 9] 10] Typically, a set of rules is used since it is the most easily implemented and understood. A rule usually has an if. then. format, where the if part is called a symptom or an evidence, and the then part is called a fault. The expert ....

D.A. Waterman, "A Guide to Expert Systems", Addison-Wesley, Reading, MA, 1986.

....the knowledge about the communication language and the communication system. Each communication sentence corresponds to a template. Each template corresponds to a procedure to interpret it. The relationship between a template and the corresponding procedure is represented by a production rule [18], such as IF template THEN procedure The IF part of each rule is a template. A template is a sequence of elements. Each element may be a constant, the symbol , the symbol followed by a variable X (e.g. X) the symbol , the symbol followed by a variable X (e.g. X) the symbol followed ....

....III) Any fourth level node is a pointer to all templates whose first three terms match the top level, the second level, and the third level nodes in this branch, respectively. IV B The inference engine of the interpreter The inference engine of the interpreter uses a forward reasoning strategy [18]. Whenever the inference engine receives a communication sentence to be interpreted, it first searches for an applicable rule then executes the procedure in the THEN part of that rule with the sentence as its parameter. The key point here is how to find an applicable rule. In order to speed up ....

D. A. Waterman, A Guide to Expert Systems, Reading, Mass.: Addison-Wesley, 1986.

.... Mitchell (1996) or Shavlik and Dietterich (1990) An alternative to the inductive learning paradigm is to build a concept description not from a set of examples, but by querying experts in the field and directly assembling a set of rules that describe the concept (i.e. build an expert system; Waterman, 1986). A problem with building expert systems is that the theories derived from interviewing the experts tend to be only approximately correct. Thus, while the expert provided domain theory is usually a good first approximation of the concept to be learned, inaccuracies are frequently exposed during ....

Waterman, D. (1986). A guide to expert systems. Addison Wesley, Reading, MA.

....part A: 62 Spoken Language Dialogue Systems Report 1 if (C satisfied) then (do A) 3. An interpreter (inference engine) that chooses and applies rules. Production formalisms are all aimed at realisation. Archetypical production formalisms are production systems (e.g. EMYCIN and HEARSAY III [Waterman 1986]) and Prolog [Kowalski 1979] ATNs (augmented transition networks) and shape grammars are other examples of production formalisms. An ATN [Woods 1970] is a recursive transition network to which are added some extra features. Shape grammars [Stiny 1980] are a special kind of production rules where ....

....logic is based on fuzzy set theory which is again based on the idea that an object x belongs to a fuzzy set A with a certain grade of membership. Uncertainty formalisms have performed reasonably, e.g. hidden Markov models dominate speech recognition, and expert systems like MYCIN and PROSPECTOR [Waterman 1986] have become well known. However, systems based on uncertainty require tedious training (Hidden Markov Models) or ad hoc assignments (MYCIN) If observations are sparse and the problem domain open, the meaning of the numbers become problematic to define. How do we convert from human terms to ....

[Article contains additional citation context not shown here]

Waterman, Donald Arthur: "A Guide to Expert Systems", The Teknowledge Series in Knowledge Engineering, Addison-Wesley, 1986.

No context found.

Waterman, D. #1986#. A Guide to Expert Systems. Addison Wesley.

No context found.

Waterman, D. A., (1986). A Guide to Expert Systems. Addison-Wesley, Reading, MA.

No context found.

Donald A. Waterman. The guide to expert systems. Addison-Wesley Longman Publishing Co., Inc. Boston, MA, USA, 1985.

No context found.

Waterman, D A A Guide to Expert Systems Addison-Wesley, USA (1986)

No context found.

Waterman, D.A. (1986). A Guide to Expert Systems. Reading Mass: Addison Wesley.

No context found.

Waterman, D. A. A Guide to Expert Systems. Addison Wesley Publishing Co., 1986.

No context found.

D.A. Waterman (Ed.). (1986). A Guide to Expert Systems. Addison-Wesley.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC